[REDACTED] 2024 | Decentralizing the Super App

Captions as YouTube publishes them, usually machine-generated, so product and people names can be misheard.

1. 0:00

   I'm I'm L I'm I'm super excited to to be here today um we've heard a lot of really excellent talks over the past few hours uh you know discussing the technical social progress stuff happening in in the near ecosystem and the broader crypto ecosystem um I want to zoom out a little bit and kind of consider the big picture uh I didn't know I was going to be on stage you know two slots after Edward sden but it

2. 0:24

   actually works out perfectly Ed and I had a little huddle before this and we coordinated our talks now I'm joking but um there's going to be some overlap between some of the ideas that he discussed and some of what I'm going to be talking about here um which is decentralizing the super app right so big picture applications we're using every day why aren't they decentralized um I am Lane as I said I am a former ethereum core developer and

3. 0:48

   for the past few years I've been um building a protocol called space mesh a layer one um been a lurker in the near ecosystem all along um and excited to to to get more Hands-On with near in the coming year um and I think the most important thing to share is I'm an application developer that's what I was in a former life um but I've been focused on infrastructure these past seven almost eight years um all along wanting to get back to building applications but feeling that like the infrastructure that we needed to do it

4. 1:20

   wasn't there I want to build like really boring unsexy things I'm the opposite of a defi developer I'm the opposite of an nft developer I want to build boring stuff like messaging apps chat calendaring things like that that's what I've wanted to work on all along um I'm going to talk about why it's so hard to build usable decentralized applications today uh about what the experience of building these applications is like uh

5. 1:45

   about some of the things that we need to continue to uh improve to make it easier to build de centerless applications and we'll bring it home by looking at how this all fits in with ideas like user owned internet and user owned AI so let's start with the obvious uh application telegram I'm sure everyone uh listening to this talk is intimately familiar with the app like many of you I

6. 2:09

   use this application every day it's kind of replaced email for me to a large extent some of you may actually be using it right now as we speak um but I'd ask you to reflect for a second how decentralized is this application to what extent do you have sovereignty over your actions your data in Telegram and and just to be clear I'm picking on telegram as an obvious example but this you could say the same thing of of all the other centralized apps we use every

7. 2:32

   day um whoa what's going on sorry very very gentle so um what really got me thinking about this was this image that I saw in August and of course Snowden spoke about this as well right this is paval durov the the CEO and founder of telegram when he was arrested back in August in France for a laundry list of reasons but basically failing to cooperate with like government demands in France and in Europe up with respect to handing over access to data you know

8. 3:02

   deplatforming users things like this um and and it it it got me wondering like why isn't telegram decentralized right don't we have the tools we need to build applications like Telegram in a decentralized way and isn't telegram like the perfect candidate for this right because if telegram were decentralized there wouldn't be a CEO to arrest and lock in a hotel room as Snowden said so just it's kind of a thought experiment and I I I wrote about

9. 3:27

   this um in my my substack so in a nutshell the problem is right the reason that Telegram and apps like it are not decentralized it's just too still too darn hard to build decentralized applications today especially ones that do things other than moving coins and tokens around right so let's drill into this a little bit does anybody recognize this image I mean obviously ethereum right this is actually a old screen grab

10. 3:54

    of the ethereum landing page from 2016 okay and I I highlighted this because of the Unstoppable applications thing in the corner here right so this is a thing that we've been speaking about for at least 8 years in ethereum and in other uh other ecosystems um why aren't we building decentralized applications I mean we have sorry why aren't we building Unstoppable applications I mean again we have some right we have some amazing defi applications uh we have prediction

11. 4:18

    markets things like this but like uh why aren't things like telegram you know being built in an Unstoppable fashion and the answer has to do with something else Edwards Snowden mentioned which is uh I think he he referred to like you have a physical body and as long as that's the case and as long as you're you have a physical body and a physical jurisdiction right there's there's a person to be locked in a hotel room right and the way he said it was

12. 4:44

    like we need to prepare ourselves uh for for a future where that's no longer the case um it it comes down to infrastructure okay there's physical servers in known locations there's companies and offices right if you want to build and operate an internet scale Global usable scalable application today you're probably not doing that today in a decentralized fashion you're probably

13. 5:07

    doing it with a company that's registered in a jurisdiction right with physical offices physical data centers you can pick a friendly jurisdiction like telegram did with Dubai but then what happens when you travel right which is what happened to our friend uh Mr durov I think it's good to for inspiration look back to kind of the reason we're all here in the first place right so um the cypherpunk mentality the the the culture out of which Bitcoin and ethereum and and all

14. 5:39

    of these other projects arose um and uh one of my favorite writers on this topic was John Perry Barlo some of you may be familiar with him um he was a poet among many other things and this comes from the Declaration of the independence of cyberspace which if you haven't read it you should read it immediately and if you have you should read it again it's short and it's extraordinar powerful and I think it does a better job than

15. 6:04

    anything else I've ever seen of really boiling down the real reasons we're here and certainly the reason I'm here governments of the industrial World You Weary Giants of Flesh and steel I come from cyberspace the new home of mind on behalf of the future I ask you of the past to leave us alone you're not welcome Among Us you have no sovereignty Where We Gather and just to be clear the idea

16. 6:27

    here is not hiding it's not that we um it's not that there's no regulation right it's that the systems of the past don't map cleanly into the things that we're building right so the question is kind of how do we get there so let's reflect a little bit on on where we are today and what it's like to build and use um decentralized applications today so how many of you in this room how many

17. 6:52

    of you listening to this talk use decentralized applications daps on a regular basis raise your hand okay yeah that's not bad that's more than half um how many of you have used a decentralized application in the past 24 hours I guess Tech technically the the the ticket app the app that we're using here for this event is is a decentralized application so that's really really cool um but how many of us use

18. 7:16

    applications that are not just Defi and not just uh you know moving tokens around things like that right obviously this is this is a popular uh dap on ethereum and many other chains as a developer one of the challenges today is that there's a whole lot of chains out there right we have layer ones we have layer twos we have side chains we have rollups we have all these things and uh it's really a headache to develop an

19. 7:39

    application and roll it out across many many chains it's bad enough just rolling out an application across all the evm chains to say nothing of the challenge of doing it on chains that have like heterogenous VMS and there's more and more of them the other thing is that the chains are still very very crowded today um you know we've definitely seen improvements here in terms of transaction throughput but I did a back of the envelope calculation and if you were to try to run telegram on ethereum

20. 8:10

    today every user would be able to send two transactions a year two transactions a year and they'd be expensive okay so the point is a 10x 100x Improvement right that's a great start but like really we need not two or three orders of magnitude we need like five six seven orders of magnitude um Improvement and throughput to be able to build even a single internet scale application on chain so let's talk about some of the

21. 8:35

    missing pieces of the puzzle uh we touched upon this one already this is just infrastructure okay to the extent possible right if we're able to increase the throughput if we're able to build better Bridges um ways to kind of move you know data and value uh from one ecosystem to another that are trustless ideally right that's a huge step in the right direction and progress is obviously being made

22. 8:57

    here uh this is this is applications so um we need applications we need more applications we need really boring unsexy applications the things I talked about right how crazy is it that my day-to-day communication with my team with my friends with my family right all of this is still sadly flowing through like very centralized uh applications um you know when the iPhone rolled out when the App

23. 9:22

    Store came out in 2007 as a platform as a powerful new platform I think people don't appreciate that one of the reasons it was so successful is because it's shipped with a suite of again really boring really unsexy proof of concept applications that work really really well right this is something again the kind of thing that I really want to work on and I hope many of you do as well uh the next topic is funding if you speak to any application developer in the room

24. 9:46

    today they will all tell you the same thing it's really really really hard to raise money to build applications as opposed to infrastructure It's relatively easy to raise money to build infrastructure because of tokens and it's sort of a tried and true business model in the industry VCS are excited to fund infrastructure uh and a big theme lately has been where are all the apps right it's more infrastructure more infrastructure um obviously ecosystem funds grants things like this can help

25. 10:10

    and again we're making some progress here but it's still too hard today and and all of that together it's still not enough we still need something else and um it's really hard I'm trying to kind of put my finger on on exactly what to call it but for lack of a better word I'm going to call it a substrate right so like we need to just remove complexity from the user layer remove complexity from the developer experience

26. 10:35

    um we need like platforms a substrate that kind of ties all of these pieces together and I think the best example of this is identity okay so this is identity um think about the number of identities that we all have to manage every day and the forms they come in right from from government IDs to phone numbers email addresses uh telephone numbers SMS codes we have to receive uh cryptographic key pairs obviously your

27. 10:58

    Facebook your Google login I mean it's just it's absurd how many we have um and unless we're until we're able to boil away some of this complexity and build applications that just work uh we're not going to achieve the goal of building something like like a super app that's decentralized so that brings us to use our own internet and bringing things home to kind of you know talk about some of the things that are happening in near and how I think these are um critical

28. 11:21

    components on the road to achieving Mass adoption of usable decentralized applications there's many many names for this I use the word substrate a moment ago but actually I think user own internet is like the best encapsulation I've ever heard of this idea and of course there's different pieces to the puzzle um so chain abstraction is the first one that comes to mind I I don't know how I ended up with with this this image for chain abstraction um but again

29. 11:47

    like as a as a frequent user of decentralized applications it's quite frankly ridiculous how many different wallets I'm I'm using every day and how many different um you know uh accounts I'm managing in each of those wallets um chain abstraction is the only Vision I've seen of of how we can escape this kind of zero some mindset um and and really achieve the level of Simplicity in ux that we need to to build applications that just work and the Holy

30. 12:11

    Grail the direction we're all going ultimately is of course Ai and and agents that are able to act on our behalf it's it's a Hot Topic right now for a reason but we need AI agents that know everything about us right the only way these are going to be useful is if they have access to our data if we trust them if they're running locally on our own device if they're ideally running inference against models that are open

31. 12:35

    source models Community govern models things like this that's the only way this is going to work it's the only way that we're going to effectively out telegram telegram um and I think that that the near vision for not only use our own internet but use your own AI is is the again the necessary step on that path if you saw Ila's uh keynote here this morning you will have seen this slide and I just thought he so perfectly captured like like these are the pieces

32. 12:57

    of puzzle that are rapidly falling into place and these are literally the solutions we need to the problems I was just outlining I just spoke about user own AI I talked about chain abstraction fixing a lot of the UI problems um and sharded blockchain so just to zoom in on this for a second um Bowen put this slide up 86 billion transactions a day that's mind-blowing and I have to say if this came from anyone else from any other project I would be calling BS and saying like Okay this is just some crazy

33. 13:25

    marketing number but um but I do think we're not far from this being a reality and that is the order of magnitude right I said we need like four five six orders of magnitude increase in throughput to build even a single global scale application like telegram um I think we'll be there very very soon so look I like big hairy audacious goals I think that as a community as an ecosystem our

34. 13:49

    goal should be one year from now now that all of these pieces are coming together we can not exactly decentralize telegram again telegram is just a is an example here but build a thing that is a super app that that you know allows us to shift a lot of the the Reliance we have on applications centralized applications like telegram um onto something better that we can conceive of and build together in the near ecosystem thank you